Autism spectrum disorders are defined by deficits of social-communication and the presence of restrictive and/or repetitive behaviors. Recent epidemiologic studies have documented an increase in the number of children identified with autism spectrum disorder (ASD) over the past decade and according to some, the current numbers indicate that as many as 1% of children have an ASD. The life-long impairments in communication and social function are often complicated by the presence of medical comorbidities, including epilepsy, (and epileptiform discharges), gastrointestinal disturbances and sleep disorders. Little is known about the pathophysiology of these comorbid conditions and even less about treatment of these disorders in autistic individuals. A variety of traditional medical and alternative biomedical approaches have been tried and anecdotally reported to be useful for one or a few individuals, but none has proven efficacious when subjected to rigorous randomized, placebo-controlled investigation. Thus, identifying the etiology, pathophysiology and treatment of the medical comorbidities of autism is an important goal for research. When multiplied by the millions of children reported to be affected by ASD, the potential public health impact is tremendous. Sleep disorders in ASD are of particular interest to our research group and can be reliably investigated using polysomnography (PSG), a non-invasive recording of a variety of sleep parameters. Preliminary observations from another of our projects, Clinical and Behavioral Phenotyping of Autism and Related Disorders, (NCT00298246) revealed that children with autism spent an abnormally short time in the Rapid Eye Movement (REM) stage of sleep compared to total sleep time and had a prolonged latency to REM sleep. Rapid Eye Movement sleep is thought to play a key role in learning and memory and may be particularly important for the integration of emotional memories in learning and behavior. Although its relationship to human neurodevelopment is unknown, animal studies have shown that REM sleep increases after intensive learning sessions. These laboratory findings formed the basis for the hypothesis that REM sleep is important for the healthy development of the central nervous system and may also be a useful indicator of brain plasticity. A small open-label pilot study of donepezil, a reversible inhibitor of acetylcholinesterase, was undertaken to evaluate its ability to enhance REM sleep in 2 to 11 year-old children with autism and a relative REM deficiency(defined as below 2 standard deviations of observed normative data for age). The primary purpose of the study was to determine the minimum dosage of donepezil required to normalize the amount of REM sleep. The side-effects profile of donepezil was also of interest in this pilot study, as there has been little previous experience with pediatric administration of the drug. Results of this study showed that 2.5 mg per day of donepezil was sufficient to significantly increase REM sleep into the normal range, as a percentage of total sleep in children with autism. The drug was also shown to decrease REM latency. The pilot data provided support for a larger, placebo-controlled study which has just been launched to investigate the use of donepezil in the treatment of ASD, via normalizing REM with the goal of increasing the potential for learning and plasticity(NCT01887132). Enrollment in the trial is underway, with baseline PSG's determining whether subjects are randomized to receive either donepezil or placebo (if REM sleep deficits are present) or receive open-label donepezil (to examine behavioral effects of donepezil among children without REM sleep deficits). In addition, we recently published a study regarding the methodology used for evaluating sleep architecture abnormalities in clinical populations. Although PSGs are expensive and time consuming exams, they can yield valuable information regarding sleep physiology in special populations. We compared first and second night polysomnography (PSG) results in order to test the so-called first night effect among children with autism. This was one of the first studies to examine night-to-night variation of sleep parameters among children with autism or other developmental disorders. The goal of the analysis was to examine sleep quality on the first and second nights that PSGs were obtained from 16 well-characterized children with ASD. Importantly, our results suggest that only a single night in laboratory is needed to detect abnormalities of sleep architecture in this population. Not only is this a potentially large cost savings, but it also improves the tolerability and feasibility of the sleep studies for children with autism.